The enzymatic activity of AOC3 (amine oxidase, copper containing 3; vascular adhesion protein 1) has been described already in 1967 as a monoamine oxidase activity in the plasma of chronic liver disease patients (Gressner, A. M. et al., 1982, J. Clin. Chem. Clin. Biochem. 20: 509-514; McEwen, C. M., Jr. et al., 1967, J. Lab Clin. Med. 70: 36-47). AOC3 has two closely homologous genes in the human genome: AOC1 which corresponds to a diamine oxidase (Chassande, O. et al., 1994, J. Biol. Chem. 269: 14484-14489) and AOC2, a SSAO with a specific expression in the retina (Imamura, Y. et al., 1997, Genomics 40: 277-283). AOC4 is a sequence that does not lead to a functional gene product in humans due to an internal stop-codon (Schwelberger, H. G., 2007, J. Neural Transm. 114: 757-762).
The enzyme contains an oxidized 2,4,5-trihydroxy-phenalalanyl-quinone (TPQ) and a copper ion in the active side. This characteristic catalytic center classifies the semi-carbazide-sensitive amine oxidase (SSAO, copper-containing amine:oxygen oxido-reductase (deaminating)): The type II membrane protein belongs to the family of copper containing amine oxidases together with several other diamine and the lysyl oxidases. However the later enzymes can be distinguished from AOC3 in their preference for diamines and the low sensitivity towards semicarbazide inhibition (Dunkel, P. et al., 2008, Curr. Med. Chem. 15: 1827-1839). On the other hand monoamine oxidases contain the flavin adenine dinucleotide (FAD) cofactor in their reactive center like the monoamine oxidase A (MAO-A) and MAO-B and follow therefore a completely different reaction scheme.
AOC3 catalyzes a two-step reaction mechanism for the oxidative deamination of primary aliphatic and aromatic amines. In a first reaction the primary amine forms a Schiffbase with the TPQ aldehyde. This covalent bond is hydrolyzed, releasing the aldehyde product and a substituted TPQ residue in the active core. In the presence of oxygen, TPQ is oxidized under the formation of ammonia and peroxide with the help of the copper molecule (Mure, M. et al., 2002, Biochemistry 41: 9269-9278). Several substrates of AOC3 have been described, like the physiological amines methylamine, dopamine, or aminoacetone, whose products of oxidation have been associated to cardiovascular pathologies (Yu, P. H. et al., 1993, Diabetes 42: 594-603). Synthetic amines have been optimized for their turnover by AOC3 like benzylaminde derivates (Yraola, F. et al., 2006, J. Med. Chem. 49: 6197-6208), C-Naphthalen-1-methylamine (Marti, L. et al., 2004, J. Med. Chem. 47: 4865-4874) or luciferin derivates (Valley, M. P. et al., 2006, Anal. Biochem. 359: 238-246). The later substrate can be used for the sensitive detection of AOC3 activity in plasma, tissue or for biochemical characterization of the enzyme.
Under pathophysiological conditions of high AOC3 activity the aldehyde products are highly reactive, leading to advanced glycosylation end products (Mathys, K. C. et al., 2002, Biochem. Biophys. Res. Commun. 297: 863-869) which are regarded as markers and drivers of diabetes associated inflammatory mechanisms.
Further the byproduct hydrogen peroxide is sensed by the tissue as a messenger of inflammation. The intermediate is able to activate endothelium and is fostering the activation of leukocytes.
The binding and modification of Siglec-10 as a membrane bound substrate provides a mechanistic understanding of how the enzymatic reaction could trigger the leukocyte transmigration through activated endothelia. The binding of Siglec-10 to AOC3 was shown in several adhesion assays and led to increased hydrogen peroxide production (Kivi, E. et al., 2009, Blood 114: 5385-5392). Binding of activated leukocytes to the dimeric, extracellular AOC3 via the Siglec-10 generates a transient association to the activated endothelial. Therefore the rolling velocity of leukocytes is reduced, which increases the transmigration of leukocytes into the interstitium of inflamed tissues. Further, a conserved RGD-motif on the surface of AOC3 argues for its adhesive role: The deletion of this sequence reduced leukocyte recruitment (Salmi, M. et al., 2000, Circ. Res. 86: 1245-1251) probably via a lack of integrin β1 binding activity (Aspinall, A. I. et al., 2010, Hepatology 51: 2030-2039).
This finding correlates to the phenotype of AOC3 knock out mice, which exert a reduced leukocyte and lymphocyte transmigration capacity (Stolen, C. M. et al., 2005, Immunity. 22: 105-115) into lymphoid organs and adipose tissue (Bour, S. et al., 2009, Am. J. Pathol. 174: 1075-1083).
AOC3 activity can be found in most tissues and is mainly expressed in endothelial cells, smooth muscle cells and adipocytes (Boomsma, F. et al., 2000, Comp Biochem. Physiol C. Toxicol. Pharmacol. 126: 69-78; O'Sullivan, J. et al., 2004, Neurotoxicology 25: 303-315). In humans, in contrast to mice, AOC3 activity is constitutive in the liver sinusoideal endothelial cells (McNab, G. et al., 1996, Gastroenterology 110: 522-528) and mRNA expression is further upregulated under inflammatory conditions in this tissue (Lalor, P. F. et al., 2002, Immunol. Cell Biol. 80: 52-64); Bonder, C. S. et al., 2005, Immunity. 23: 153-163). AOC3 not only exists as a membrane protein, but can also be found as soluble plasma activity probably due to a metalloprotease mediated shedding process (Abella, A. et al., 2004, Diabetologia 47: 429-438); Boomsma, F. et al., 2005, Diabetologia 48: 1002-1007; Stolen, C. M. et al., 2004, Circ. Res. 95: 50-57)). Elevated levels of soluble AOC3 have been observed in diabetes (Li, H. Y. et al., 2009, Clin. Chim. Acta 404: 149-153), obesity (Meszaros, Z. et al., 1999, Metabolism 48: 113-117; Weiss, H. G. et al., 2003, Metabolism 52: 688-692), congestive heart failure (Boomsma, F. et al., 1997, Cardiovasc. Res. 33: 387-391), end-stage renal disease (Kurkijarvi, R. et al., 2001, Eur. J. Immunol. 31: 2876-2884) and inflammatory liver disease (Kurkijarvi, R. et al., 1998, J. Immunol. 161: 1549-1557). For the later, levels of AOC3 plasma activity have been correlated to liver fibrosis and serve as a predictor in patients with NAFLD (Weston, C. J. et al., 2011, J. Neural Transm. 118: 1055-1064). After transplantation of cirrhotic livers, high AOC3 plasma levels dropped drastically, which argues for the liver as the major source of plasma AOC3 activity under this pathological condition (Boomsma, F. et al., 2003, Biochim. Biophys. Acta 1647: 48-54).
The role of AOC3 in the activation of inflammation via peroxide generation and the recruitment of leukocytes to activated endothelium makes it an attractive target for the treatment of inflammatory components in several diseases. Therefore a variety of small molecular compounds and antibodies have been tested in different disease animal models. Amongst those, the inhibition of AOC3 showed beneficial effects in the models of melanoma and lymphoma cancer (Marttila-Ichihara, F. et al., 2010, J. Immunol. 184: 3164-3173), acute and chronic joint (Tabi, T. et al., 2013, J. Neural Transm. 120: 963-967) or lung (Foot, J. S. et al., 2013, J. Pharmacol. Exp. Ther. 347: 365-374) inflammation, diabetic macular edema (Inoue, T. et al., 2013,
Bioorg. Med. Chem. 21: 1219-1233), kidney fibrosis (Wong, M. et al., 2014, Am. J. Physiol Renal Physiol 307: F908-F916), liver allograft rejection (Martelius, T. et al., 2004, Am. J. Pathol. 165: 1993-2001) and non-alcoholic liver disease.
The development of a selective, potent and well tolerated AOC3 inhibitor would therefore be beneficial for the treatment of the respective human diseases.
AOC3 inhibitors are known in the art, for example, the compounds disclosed in WO 2012/124696. The benzonitrile derivatives of the present invention may provide several advantages, such as enhanced potency, reduced plasma protein binding, improved CYP (cytochrome P450) enzyme profile and high metabolic stability, high chemical stability, improved tissue distribution, improved side effect profile and/or tolerability and in consequence low toxicity, reduced risk to cause adverse events or undesirable side effects, and enhanced solubility.